PhD student, Freie Universität Berlin
3 papers at NeurIPS 2025
we train transferable normalizing flows to sample from peptide Boltzmann distributions up to 8 residues
An energy-based diffusion model regularized with the Fokker-Planck equation to improve the consistency between iid sampling and simulation
We investigate the benefits of using path gradients to fine-tune CNFs initially trained by Flow Matching, in the setting where a target energy is known.
